The objective of this project is to evaluate medicinal drugs for possible carcinogenic effects. Screening studies will be accomplished by following up a group of 149,000 patients in a prepaid health plan who had computer-sorted pharmacy records of prescriptions dispensed from July, 1969 through August, 1973. Increastd incidence of any cancer among users of a drug as compared to expected incidence derived from the entire cohort, will indicate a drug-cancer association to be evaluated further. More definitive study of hypothesized drug-cancer relationships will be accomplished by case-control studies using the patients' longitudinal medical records.